KR20040083384A - Parallax barrier and multiple view display - Google Patents

Abstract

PURPOSE: A parallax barrier and a multiple view display seeing two or more different images on same display screen are provided to enhance visibility, reduce crosstalk between windows, improve a degree of viewing freedom, increase luminance, and improve color matching by widening a viewing angle between viewing regions. CONSTITUTION: A parallax barrier includes a plurality of substantially opaque regions defining a plurality of groups of parallel slits. Each of the groups includes N number of the slits where N is an integer greater than one. The slits of each group are spaced apart with a first pitch b1 in a direction perpendicular to the slits and the groups are spaced apart with a second pitch b2, in the direction perpendicular to the slits, greater than N.b1. The slits of each group have substantially a same maximum light transmission.

Description

Parallax barriers and multi-view display {PARALLAX BARRIER AND MULTIPLE VIEW DISPLAY}

<Related application>

This application requires 35 U.S.C. Claims No. 0306516.6, filed in the UK on March 21, 2003, under 119 (a), the contents of which are incorporated herein by reference.

The present invention relates to parallax barriers, and multiple view displays comprising such barriers. Such a display can be used when two or more others need to see two or more different images on the same display screen. The viewing angle difference is preferably quite large (about 60 degrees) between the viewers.

Such a display can also be used as a three-dimensional (3D) display, in particular as a compound autostereoscopic display. Such displays can be used for 3D mobile handsets, 3D games, 3D computer monitors, 3D laptop displays, 3D workstations, and 3D professional imaging (ie, medical, design, and architectural applications). In some 3D displays, it is necessary to increase the eye separation distance between two eyes for a specific viewing distance or to reduce the screen distance for the distance between two particular eyes. Both of these cases require a wider separation angle.

Over the years, displays have been designed for multiple users and have been optimized to allow viewers to see the same good picture quality from different angles to the display. This assumes that multiple users need the same information from the display. However, there are many applications where it is desirable for individual users to see different information from the same display. For example, in a car, a passenger may want to watch a movie, but a driver may want to see satellite navigation data. If two displays are used in this case, the driver will be able to watch the movie, which will distract the mind and the two displays will take up extra space and increase costs. In a computer game, each player may want to see the game from his point of view. This is now done by each player viewing the game from his or her own perspective on a separate screen. This takes up a lot of space and is not practical for portable games.

By showing one or more images to one or more users on one display, space and cost can be significantly reduced. This may be desirable in airplanes where each passenger is provided with his own video screen. By providing two or more passengers with a single central screen but being able to choose their own movie, the cost, space and weight can be greatly reduced. It can also prevent other users from seeing what the user sees. This may be desirable for games as well as security applications such as banking or sales transactions.

In normal vision, both human eyes perceive all things of the world by different perspectives due to their distinct positions in the head. These two perspectives are used by the brain to assess the distance to various objects in a scene. In order to construct a display that will effectively display three-dimensional images, it is necessary to reproduce this situation and provide a so-called "stereoscopic pair" of images, one for each eye of the observer.

Three-dimensional displays fall into two categories depending on the method used to provide different views to the eye. Composite parallax displays typically display both images over a wide viewing area. However, for example, each view is encoded according to the color, polarization state, or time of the display, so that the filter system of the glasses worn by the observer can separate the view and each eye can only see the view intended for that eye. Will let you see.

The compound parallax display does not require any viewing aids to be worn by the observer, but the two views are only visible from a limited area of space. The area of the space where an image can be seen throughout the display active area is called the "viewing region". If one of the viewer's two eyes is in one viewing area and the other eye is in the viewing area for the other image of the pair, then the correct view will be visible and the 3D image will be recognized.

In the case of a flat panel compound parallax display, the viewing area is typically formed by a combination of the pixel structure of the display unit and optical elements, collectively called parallax optics. An example of such optics is a parallax barrier. This device is a screen with vertically transmissive slits separated by opaque regions. This screen can be installed in front of a spatial light modulator (SLM) having a two-dimensional array of pixel apertures as shown in FIG.

The display comprises an active matrix thin film transistor (TFT) substrate (1), a counter substrate (counter substrate) 2, a pixel (pixel) surface (3) formed by a liquid crystal layer, a polarizer (4) and a viewing angle increasing film (5). A transmissive spatial light modulator in the form of a liquid crystal device (LCD) is provided. The SLM is irradiated by a backlight (not shown) which irradiates in the direction indicated by the arrow 6. This display is a front parallax barrier type and has a parallax barrier with a substrate 7, an aperture array 8 and an anti-reflection (AR) coating 9.

The SLM is aligned such that a row of pixels extending perpendicular to the normal view is provided with a horizontal pitch p. The parallax barrier provides an array 8 of openings or slits, wherein the slits are parallel to each other and extend parallel to the column of pixels. The slits have a width of 2w, a horizontal pitch of b, and are spaced apart from the pixel surface 3 by a distance s.

The display has an intended viewing distance r _{o such} that the left and right viewing windows 10, 11 are in the widest portions of the viewing area defining the window plane 12. The viewing windows 10, 11 have a pitch e that is substantially equal to the distance between two typical or average human eyes. The center of each main viewing window 10, 11 is half angle α with respect to the display normal.

The pitch b of the slits of the parallax barrier is chosen to be close to an integer multiple of the pixel pitch p of the SLM such that groups of pixel columns are associated with a particular slit of the parallax barrier. 1 shows an SLM with two columns of pixels associated with each slit of a parallax barrier.

2 shows the angular zones of light generated from the SLM and the parallax barrier with the SLM and the parallax barrier having an exact integer multiple of the pitch b of the pixel column pitch p. In this case, the angular forming regions coming across the display panel surface from different locations are mixed with each other so that there is no pure view area for Image 1 or Image 2. To solve this, the pitch b of the parallax optics is slightly reduced so that the angle forming area converges on the window plane 12 in front of the display. This change in parallax optics pitch is called " viewpoint correction ", which is shown in FIG. The viewing area thus produced is roughly kite-shaped.

For color displays, each pixel is typically provided with a filter associated with one of the three primary colors. By controlling groups of three pixels, each having a different color filter, substantially all visible colors can be made. In a compound parallax display, each compound parallax image "channel" must include a sufficient number of color filters for balanced color output. Many SLMs have color filters arranged in vertical rows due to ease of manufacture so that all pixels within a given column have the same color filter associated with them. If you use parallax optics with this SLM such that three columns of pixels are associated with each slit (or lenslet), you will only see one color in each viewing area. This can be avoided using, for example, the technique disclosed in EP 0 752 610.

The function of parallax optics is to limit the light transmitted through the pixel to any output angle. This limitation defines the viewing angle of each column of pixels behind a given slit. The angular range of the view of each pixel is determined by the refractive index n of the glasses, the pixel width p, and the distance s between the pixel and the parallax optic plane, given by the following equation.

In order to increase the angle between the viewing windows, it is necessary to increase the pixel pitch p, reduce the gap between the parallax optics and the pixel s, or increase the refractive index n of the glass. It is not easy to change any of these variables. Changing the refractive index of the substrate glass significantly is not always practical or cost effective. Pixel pitch is typically limited by the required resolution specification of the panel and thus cannot be changed. Incidentally, increasing the pixel pitch requires a similar increase in the parallax barrier pitch, which makes the barrier more visible, compromising the final picture quality. The reduction of s results in manufacturing problems associated with manufacturing and handling thin glass. Thus, it is difficult to create 3D or multi-view displays with wide viewing angles using standard parallax barriers.

One option for increasing pixel pitch and thus viewing angle is to rotate the pixel configuration such that the color subpixels extend horizontally rather than vertically, as described in JP7-28015. This increases the pixel width by three times, thus increasing the viewing angle by about three times. As mentioned above, this has the disadvantage that as the pixel pitch increases to increase visibility, the barrier pitch increases. Manufacturing and driving such non-standard panels may not be cost effective. Incidentally, there may be applications where the increase in viewing angle needs to be three times larger than the standard configuration and thus simply rotating the pixels may not be enough.

The window plane defines the optimal screen distance of the display. Observers whose eyes are in this plane will get the best performance of the display. As the viewer moves laterally in this plane, the image on the display remains until the viewer reaches the edge of the viewing area, where the entire display will quickly change to the next image as the eye moves to the adjacent viewing area. The line of the window plane in the viewing area is often referred to as the "viewing window."

In an ideal display of the type shown in FIGS. 1, 2 and 3, the intensity distribution of light across each viewing window would be a “top hat” function. In other words, for each viewing window, the light intensity will be constant across the viewing window and zero outside the viewing window of the viewing plane. However, the window intensity distribution worsens, so that the observer's lateral and longitudinal viewing freedoms are reduced compared to that shown in FIG. This is caused by diffraction through the aperture as well as the gap between the pixels, so that the edges of the windows become dark areas. In an ideal display, the right eye image data will not be present in the left eye viewing area and vice versa. In practice, however, crosstalk occurs so that each eye can see some of the light intended for the other eye.

Although a particular parallax element (slit or lenslet) is in principle associated with one group of pixel columns, adjacent groups of pixel columns will also be imaged by this element. Imaging of these groups produces lobes of the viewing area that are repeated either to the center lobe or to the zero order lobe. These lobes repeat all the properties of the central lobe, but are largely affected by the imperfections and aberrations of the optical system and will eventually become unusable as the lobe order increases. Higher order lobes may be used to achieve wider angles between the windows. However, performance will be greatly reduced.

For example, by using parallax optics with a pitch of b-3p, window A and C may be used in FIG. 4A for 3D viewing or “dual view” where window B is black or contains any data. Can be. While this provides an increased viewing angle, the luminance is reduced to two thirds of the luminance produced by the barrier with pitches b to 2p because every third pixel is "shown" in the unused window. Increasing the barrier pitch will increase the visibility of the barrier and the horizontal resolution will decrease. This window is smaller than a window of a dual view system having the same viewing angle as shown in FIG. 4B, and thus the viewing freedom will be reduced. Another problem in this example is that, if the color filters are arranged in RGB columns, only each window can see a "subpixel" of one color.

5 is the slit width Shows a parallax barrier with a pitch b to 6p apart. This results in windows at separation angles that are three times the separation angle of the standard parallax barrier of pitch b-2p. Since the slit width is 3p, all three color filter colors 15, 16, 17 are seen at the same ratio. However, this is only the case in the exact center of the viewing window as shown in FIG. 5A. As the viewer moves away from the center of the viewing window, neighboring image data 15a will be visible, and crosstalk will occur as shown in FIG. 5B.

To reduce this crosstalk, the slit width can be reduced to less than 3p. However, this results in an uneven color balance. As shown in FIG. 6A, when the viewer is in the center of the viewing window, the “white” pixels are pale green because the green subpixel 16 can be seen more than each of the red and blue subpixels 15, 17. Will look as Because of the reduced slit width, the viewer can move away from the center of the viewing window without seeing adjacent image data. However, the color balance changes with the viewing angle and the "white" pixels will appear bluish green, as shown in Figure 6b for movement in one direction.

Another way to achieve a larger separation angle is to use standard b-2p barriers separated by two window widths and two secondary lobes A and D as shown in FIG. 7A. This will have the same barrier visibility, brightness and horizontal resolution as the two viewing systems shown in FIG. 7B with the same angle. However, the viewing freedom will be greatly reduced.

Another known kind of directional display is a back parallax barrier display as shown in FIG. 8. In this case, the parallax barriers 7 and 8 lie behind SLM 1 to 5, ie between the SLM and the backlight. This configuration has the advantage that the barrier remains intact behind the SLM.

Lenticular screens are used to orient interlaced images in multiple directions, which can be designed to provide 3D images or multiple images in multiple directions. Real lenses tend to suffer from scattering and poor antireflection performance, making the surface extremely visible in both the ambient and backlight irradiated environments. Thus, the picture quality of the lenticular screen can be poor and the system suffers from similar problems as a parallax barrier such as the need for proximity to image pixels.

There is also a holographic method of image splitting, but this method suffers from viewing angle problems, pseudoscopic regions, and control difficulties of the image. Micropolarizer displays use a polarization directional light source and a high precision micropolarizer element patterned in alignment with LCD pixels. These displays offer the potential for high quality windows as well as 2D / 3D functionality in a compact package. The main requirement is to integrate the micropolarizer device into the LCD to avoid parallax problems.

US 6 424 323 discloses an image reflection system having a lenticular screen overlying a display element. This display is controlled to provide at least two independent images viewed from different viewing positions.

JP 7-28015 discloses the use of a patterned pixel shape with a lenticular barrier with a window formed by the display having a minimum crosstalk. Crosstalk is reduced by shifting the relative position of the pixels and properly aligning the spacing and orientation.

Other known types of multi-view displays are disclosed in WO 98/27451, DE 19822342 and JP H7-104212.

JP-A-8-36145 discloses a parallax barrier in which a slit has a pitch randomly selected from among a plurality of predetermined pitches. The selected pitch can be repeated as groups across the barrier.

GB-2352573 discloses a parallax barrier in which slits each having a plurality of sub-apertures are uniformly spaced apart.

According to a first aspect of the invention, in a parallax barrier having a plurality of groups of parallel slits, each group comprises N slits (N is an integer greater than 1), and the slits of each group are slits A parallax barrier is provided that is spaced apart at a first pitch b1 in a direction perpendicular to and spaced apart in a direction perpendicular to the slit at a second pitch b2 greater than N · b1.

The maximum transmission of light through each slit in the group may be substantially the same.

Adjacent slits may be separated by a barrier portion having a defined width.

The second pitch b2 may be substantially equal to 2 · N · b1.

In one embodiment, N is two. In another embodiment, N is 3.

The barrier may have an active element that provides slit groups in the first mode of operation and an alternative arrangement of slits in the second mode of operation. The slits may be spaced at a substantially uniform pitch in a direction perpendicular to the slits in the second mode. The barrier may have a third mode of operation in which the barrier exhibits a substantially uniform transmission for light over its operating area.

According to a second aspect of the invention, there is provided a multi-view display having a barrier and a spatial light modulator according to the first aspect of the invention.

The modulator may have a plurality of pixel columns extending parallel to the slots.

The pixel columns may have a third pitch p in a direction perpendicular to the longitudinal direction of the column, different from the first pitch, to provide viewpoint correction. The first pitch b1 can be given by the following equation.

Where p is the pitch of the columns and e is the pitch of the main viewing window generated by the display.

The rows may have a third pitch p in a direction perpendicular to the longitudinal direction of the columns that is greater than the first pitch.

The columns may have red, green and blue columns.

The columns may be arranged as repetitive groups, each group arranged in blue, red, blue, red, green, blue, green, blue, red, green, red and green order.

The columns can be arranged as repeating groups, with each group arranged in the order green, green, blue, blue, red and red.

The columns can be arranged as eighteen repetitive groups, each group containing three consecutive pairs of the same triplet, the order of the colors of the triplets of the consecutive pairs being cycled by one position relative to each other.

The columns can be arranged as 36 repeating groups, each group containing six consecutive pairs of the same triplet, with the triplets of consecutive pairs having an order that includes all permutations of red, green, and blue.

The display may include a display driver for supplying an image signal to the modulator representing the plurality of views as interlaced columns. The image signal may represent two views. The display may include a compound parallax display in which the image signal represents at least a pair of stereoscopic views.

Thus, it is possible to provide a parallax barrier suitable for use in multiple view displays having a relatively wide viewing angle between viewing areas. Different embodiments of such a display can exhibit different advantages among the reduced barrier visibility, reduced crosstalk between windows, improved viewing freedom, increased brightness and improved color matching.

1 is a schematic horizontal cross-sectional view of a multi-view display of known type.

FIG. 2 is a schematic plan view showing an angular viewing region generated by the display of FIG. 1. FIG.

FIG. 3 is a schematic top view illustrating viewpoint correction in the display of FIG. 1. FIG.

4A and 4B are schematic views of two comparative examples of a multi-view display.

5A and 5B illustrate the visibility of pixel columns for on-centre viewing and off-centre viewing, respectively.

FIG. 6 is a view similar to FIG. 5 showing a parallax barrier of reduced slit width; FIG.

FIG. 7 is a view similar to FIG. 4, showing a further comparative example of a multi-view display.

8 is a schematic cross-sectional top view of a rear parallax barrier multi view display of known type.

9 (a) and 9 (b) are schematic cross-sectional plan views of parallax barriers and displays and schematic front views of parallax barriers constituting an embodiment of the present invention.

10 is a schematic plan view of the display of FIG. 9 showing creation of a viewing area.

FIG. 11 is a view similar to FIG. 10, showing another embodiment of the present invention. FIG.

12A and 12B show the visibility of pixel columns in the display shown in FIG. 11 for on-center viewing and off-center viewing, respectively.

13A and 13B show a variant of the display shown in FIGS. 11 and 12, showing pixel column visibility for on-center viewing and off-center viewing, respectively.

FIG. 14 shows a color subpixel arrangement on a panel in left and right views for a display of the type shown in FIG. 9, with three different orders of color subpixels. FIG.

FIG. 15 shows a view similar to FIG. 14, for a display of the type shown in FIG. 11; FIG.

FIG. 16 illustrates a parallax barrier and associated pixel color filter arrangement constituting an embodiment of the present invention. FIG.

Like reference numerals in the drawings indicate like parts.

<Explanation of symbols for the main parts of the drawings>

1, 7: substrate

2: counterpart substrate

3: pixel side

4: polarizer

8: opening array

9: anti-reflective coating

21: parallax barrier

25: display driver

The display shown in FIG. 9 is a two-view directional display that can be used as a compound parallax 3D display or a display for providing two different views to irrelevant viewers. This display is provided with a spatial light modulator in the form of an LCD 20 of the type shown for example in figures 1 to 5. The LCD 20 is pixelated and operates in a transmissive mode to modulate the light passing through the pixels from a backlight (not shown). However, other types of displays may be used to modulate the light in a transmissive or reflective mode or to generate light within the display element itself (in the case of a front parallax barrier configuration).

This display has a parallax barrier 21 disposed in front of the LCD 20, ie between the LCD 20 and the observer (s). Barrier 21 may, for example, use standard emulsion techniques, or retarder or polarizer techniques, or areas such as 22 and 23 that do not substantially transmit light from LCD 20 and light from LCD 20. It may be implemented in any suitable way to provide a slit between said areas that are substantially transmissive. Regions 22 and 23 have a finite width and all slits have the same maximum light transmission.

The LCD 20 is pixelated to form rows of pixels having a pitch p that is substantially uniform in the direction perpendicular to the longitudinal axis of the column of pixels that is generally horizontal in the normal use of the display.

The slits of the barrier 21 are arranged aperiodically and extend parallel to the vertical axis of the pixel column. In particular, the slits are arranged in the form of groups of evenly spaced slits with the slits in each group evenly spaced. In the illustrated embodiment, each group is two slits similar to the pitch p of the column of pixels but spaced apart from the slits in each group by a different pitch b1 to provide viewpoint correction as described above. It is provided. Thus, the pitch b1 is somewhat smaller than the pitch p for the front parallax barrier display of the type shown in FIG. 9 but somewhat larger than the pixel column pitch p for the rear parallax barrier display of the type shown in FIG. In particular, in the case of a front parallax barrier the pitch b1 is p / (1+ (p / e)), whereas in the case of a back parallax barrier the pitch b1 is given by p / (1- (p / e)).

The groups of all slits have the same number of slits having the same pitch b1, and in the illustrated embodiment these slits extend throughout the height of the parallax barrier 21. The groups of slits are spaced in the direction perpendicular to the slits (horizontal during normal use of the display) by the same pitch b2 which is approximately four times the pitch b1 and thus approximately four times the pitch p of the pixel rows.

The display is driven by a display driver, shown schematically at 25 in FIG. 9 so that the image data for the two views to be displayed are interlaced as vertical stripes. Display driver 25 may be arranged to receive an image for display and interlace the data so that individual pixel columns display an accurate vertical slice of the images. The display driver 25 may be formed as part of a display or may be implemented partly or wholly in another device such as a computer or a microprocessor. The images may be captured “real” images or computer-generated images. The images may form a binocular parallax pair for composite parallax 3D use of the display, or display a completely different viewpoint image of a two-player game, or "movie" to one observer and satellite navigation to another observer, such as a motorist. It can be binocularly disparate, such as a series of cinematographic images for presenting an image.

9 and 10, the slits of the barrier 21 are aligned along or adjacent to the middle line of the pixel column. In practice, the slits are offset from this intermediate line alignment by a small amount to provide viewpoint correction. The display driver 25 supplies the vertical image slices 1 to 4 in FIG. 10 to the group of four pixel columns closest to each group of slits. The slits of the barrier 21 interact with the pixellation of the LCD 20 to define or create a viewing area, labeled A through E in FIG. 10. In each viewing area, each slit group limits the visibility of the column of pixels so that only two adjacent pixel columns can be seen by an observer viewing the display from the viewing area. Thus, in the zero order lobe viewing region C, only those columns displaying image slices 2 and 3 can be seen. Only pixel columns displaying image slices 1 and 2 are visible in the positive primary lobe viewing area D, whereas only pixel columns displaying image slices 3 and 4 are visible in the negative primary lobe viewing area B. Similarly, in the positive and negative second order lobe viewing areas E and A, only the pixel columns displaying image slices 1 and 4 are visible.

The display driver 25 supplies the pixel image data to the LCD 20 such that image slices 1 and 2 are provided from one of the images and image slices 3 and 4 are provided from the other images in the image. Thus, the first and second images forming the first and second views can be seen in viewing areas D and B, respectively. When providing a compound parallax view, assuming that the observer's right and left eyes are present in the viewing areas B and D, respectively, the binocular parallax pairs of the image can be accurately projected to provide a 3D effect. Conversely, the observer whose eyes are in the viewing area D can see one of the images but not the other image, whereas the observer whose eyes are in the viewing area B can see the other image but the first image Can't see

The half angle a with respect to the center of the viewing areas B and D in the display is as follows.

α = sin-1-1

Compared to conventional displays of the type shown in FIGS. 1-3 and 4A, the angle between views is about twice as high. In the case of a compound parallax display, this allows the optimal screen distance or viewing window plane to be about half the distance from the display as compared to conventional configurations. Conversely, in the case of "unrelated" views, the angle between the views for irrelevant images actually increases.

The viewing area for each side of areas B and D actually used comprises 50% of each image. Thus, since 50% of the stray light comes from the same image source, the contribution to crosstalk from adjacent viewing areas is reduced. Compared to the three views shown in FIG. 4A, the display shown in FIGS. 9 and 10 uses 50% of the available light, and each image is 50% of the pixels so that the horizontal resolution is 50% of the LCD resolution. Is displayed. For the three view displays of FIG. 4A, 33% of the light is used and 33% of the LCD resolution is used for each image.

The display shown in FIG. 11 differs from that shown in FIGS. 9 and 10 in terms of the configuration of the image slice displayed by the parallax barrier 21 and the LCD 20. In this embodiment, barrier 21 comprises a group of three slits in which each group of slits has the same pitch b1 as in the group of slits in the display shown in FIGS. 9 and 10. However, the groups of slits are spaced by a horizontal pitch b2, which is six times the pitch b1 within the viewpoint correction and about six times the pitch p of the pixel row. In addition, each slit is aligned (within viewpoint correction) along the boundary between adjacent pairs of each of the pixel rows of the LCD 20.

Each group of pixel columns associated with each group of slots in the barrier 21 displays a vertical image or view slice, represented by 1 through 6 in FIG. 11. The configuration of the barrier 21 against the column of pixels of the LCD 20 will produce a viewing area, labeled A through G in FIG. In the primary viewing area D, pixel columns displaying view slices 3, 4, and 5 are visible, whereas in viewing area E, pixel columns displaying view slices 2, 3, and 4 are visible. In viewing area C, pixel columns displaying view slices 4, 5, and 6 can be seen, while in viewing area F, pixel columns displaying view slices 1, 2, and 3 can be seen. In viewing area B, pixel columns displaying view slices 5, 6, and 1 can be seen, while in viewing area G, pixel columns displaying view slices 6, 1, and 2 are visible. The display driver 25 supplies the image data to the LCD 20 such that the view slices 4, 5, and 6 are the first images while the view slices 1, 2, and 3 are the second images. Thus, in the viewing area C, the first image is visible and the second image is virtually invisible, while in the viewing area F the second image is visible and the first image is virtually invisible. The half angle α for the viewing areas C and F is given as follows.

α = sin -1-1

This provides angle between views which is about three times the angle for the display shown in FIG. 4A. Thus, for compound parallax display purposes, the optimal screen distance or viewing window plane is much closer to the display. In other words, in order to see an unrelated image, a larger separation distance is provided for the viewing area.

The window for either side of each window used for viewing in the display of FIG. 11 contains 66.7% of the same image data. Thus, since 66.7% of the stray light comes from the same image source, the contribution to crosstalk from adjacent windows is reduced. 50% of the LCD resolution is used for each image and 50% of the available light is used for each image.

As shown in FIG. 11, by using pixel columns displaying view slices 4, 5 and 6 for the first image and view slices 1, 2 and 3 for the second image, respectively, to display red, green and blue data. , Color balance is maintained. On-centre viewing of the red, green and blue pixel columns 15, 16 and 17, respectively, is shown in Fig. 12A. However, the main cause of crosstalk occurs because of the outer edges because the outer edges of each group of slits are closest to the column of pixels having opposing image data. When the observer moves away from the center of the viewing window, adjacent image data can be viewed and crosstalk occurs as shown in Fig. 12B.

As shown in FIG. 13, crosstalk can be reduced by reducing the width of the slit and reducing the slit pitch within each group of slits. On-center viewing and off-center viewing are shown in FIGS. 13A and 13B, which correspond to FIGS. 12A and 12B, respectively. Although the reduced slit width reduces the light output of the display, the crosstalk effect is reduced while allowing greater lateral movement freedom for each observer. Although the pitch of the slits in each group in the configuration of FIG. 13 is reduced compared to the configuration of FIG. 12, the pitch of the group of slits does not change.

14A illustrates the use of a conventional type of LCD 20 in which "white" pixels are divided into repeating groups of color subpixels. In particular, the pixel columns of each group of three columns are such that all color subpixels within each column display the same color and adjacent pairs of columns repeat the red (R), green (G) and blue (B) patterns across the display. Red, green and blue filter strips are provided to display different colors. As shown for the right and left views, even if the correct color balance is obtained for this configuration, there is a substantial non-uniformity in the spacing of a single color for each view. In particular, as shown well in FIG. 14A, the spacing of the red subpixels alternates with nine and three color subpixels. Such nonuniform spacing is particularly visible in low resolution displays, thus degrading the picture quality. Also, for each view, the ordering of color subpixels does not follow the same repeating pattern of three color subpixels. For example, in the right view, the first white pixel consists of R, G, and B subpixels, but the actual arrangement is an R, G, G, and B color subpixel. Thus, one of the G subpixels belongs to the other white pixel, and this crossing in the order of the components of each white pixel can be an artifact with an undesirable image.

To avoid or reduce this problem, the color of the pixel columns may differ from the standard repeating RGB configuration, an example of an alternative configuration as shown in FIGS. 14B and 14C. In Fig. 14B, the color order of the columns is changed to the repeating pattern of GBBRBRGBGBRGRG. Thus, the color subpixels for each view are aligned as a repeating group of RGB subpixels such that no intersections exist in the arrangement of the white pixel components. In addition, the spacing of the individual color subpixels in each view alternates with 5 and 7 subpixels, which improves the uniformity of the spacing, i.e., image quality.

The disadvantage of this configuration is that if the display is required to operate in the 2D mode, the subpixel arrangement is returned to the actual LCD 20 arrangement and is no longer aligned with a repeating group of three color subpixels. will be. This results in some intersection of the arrangement of the white pixel components, resulting in an undesirable image. However, such a compromise may be acceptable because the spatial spatial resolution of the LCD is used in the 2D mode and this helps to reduce the visibility with cross effects.

Another advantage of the configuration shown in FIG. 14B is achieved by the application of such a compound parallax 3D. In this case, a 3D pixel with zero disparity, ie appearing in the same position of the display for both the left and right eyes, has only two subpixels of residual disparity. This shows that the residual disparity is improved over the configuration shown in FIG. 14A, which is six subpixels appearing alternately in the same direction or back and forth of the screen.

Another configuration of color subpixels is shown in FIG. 14C. In this case, each color component is repeated so that the columns are arranged as a pair where each pair displays the same color component. This configuration avoids intersection results in multiple view modes and provides more uniform spacing of color components for the two views. When used in the 2D mode, the display is used to have half the spatial resolution of the LCD 20, but there is no intersection of the components of the white pixels. Such a configuration can be used, for example, in displays with high resolution or in dual view displays where the native resolution of the LCD can be increased to provide high resolution images to each viewer.

In the embodiment shown in Figs. 12 and 13, the arrangement of colors is the same as that of the conventional RGB LCD 20 as shown in Fig. 15A. Avoid undesired color results seen with a single wide slit having a pitch substantially equal to a six pixel column pitch. However, if an error occurs in the pitch, width or spacing of such slit during manufacture, some color results will remain. In the configuration shown in Fig. 15A, if the contribution ratio of each group of three adjacent RGB pixel columns is not the same, white balance is not obtained. If this error occurs in each group of RGB pixel columns, this error will spread to the entire display.

An alternative order of color subpixels to avoid or reduce the impact of this problem is shown in FIG. 15B. In such a configuration, the order of the color subpixels is "rolled" by one position for each group of subpixels in each view. This pattern is repeated every three groups so that when the LCD 20 has a sufficiently high resolution, any errors are averaged across the panels, making them less or less visible to the viewer. However, the spacing of the individual color subpixels in each view varies with the pattern of 5, 5, and 8 pixels, which may result in undesirable image results.

Another configuration is shown in FIG. 15C to overcome or reduce the effect of error and to provide a more uniform color subpixel spacing. In this configuration, the order of two adjacent color subpixels is swapped, while the third color subpixel is kept in the same position. This pattern is repeated every six groups. Thus, even if the error is not averaged effectively as in the configuration shown in Fig. 15B, the spacing of the individual color subpixels will change according to the pattern of 7, 7, 6, 5, 5, 6 pixels and will therefore be less visible.

In general, a multiview display for providing a compound parallax perceptual display of binocular parallax images provides angular separation between views of about 5-10 degrees. However, for multi-view displays intended to provide different views to different observers, such as dual view displays, much larger angular spacing is typically required and typically has a value between 50 and 100 degrees. 16 illustrates a configuration that can provide such a wide variety of separation angles in various modes of operation. Such a configuration may be arranged to provide a single view or 2D mode of operation. Switching between modes can be performed electronically.

In the arrangement shown in FIG. 16, the parallax barrier 21 is of active or controllable type and has, for example, a liquid crystal element with suitable control electrode patterning and a suitable control arrangement for selecting various modes of operation. The patterning shown in FIG. 16 can be used as the patterning of control electrodes in which the same shading area or hatching area can be controlled by the same control signal.

If a 2D mode of operation is required, then the parallax barrier liquid crystal display is operated such that the entire active area is transmissive and substantially uniformly transmissive throughout the device. Thus, the barrier structure is substantially invisible and the display can be operated to provide a single resolution mode of full resolution.

If the display is required to provide a relatively small angular spacing, for example in the compound parallax 3D mode of operation, the electrodes corresponding to the areas 30 and 31 are transmissive while these areas are transparent. The remaining portion is controlled to be substantially impermeable. The barrier 21 thus acts as a conventional periodic barrier in which the slits are spaced by a substantially constant pitch over the entire barrier.

If the display needs to operate with a wider angular separation between views, the electrodes corresponding to regions 31 and 32 are such that these regions are transparent while the rest of the barrier is substantially impermeable. Controlled. The barrier then acts as a non-period barrier in the same manner as described above in this embodiment of the invention.

If a single view mode is not needed, an area such as 35 marked black in FIG. 16 is permanently impermeable and does not need to have any control arrangement.

Although narrow and wide separation angle modes have been cited according to their use in complex parallax and non-complex parallax applications, respectively, this is not essential and these modes can also be used for non-complex parallax and complex parallax operations respectively. . In addition, although the barrier 21 shown in FIG. 16 provides three slots in each group in a wider separation angle mode, each group may have a different number of slits. For example, by providing suitable control connections to the electrodes 31 and 32, it will be possible to operate a barrier with two slits in each group.

The upper part of FIG. 16 shows the color filtering arrangement for the column of pixels in the associated liquid crystal element. These columns are arranged as repeating groups of red, green and blue pixels 15, 16 and 17, respectively. Dotted lines like 38 show the relative configuration of the controllable slots of the barrier 21 with rows of pixels.

According to the present invention, there is provided a parallax barrier suitable for use in multiple view displays having a relatively wide viewing angle between viewing areas. In addition, the present invention has the advantages of reduced barrier visibility, reduced crosstalk between windows, improved viewing freedom, increased brightness and improved color matching.

Claims (24)

A plurality of substantially impermeable regions defining a plurality of groups of parallel slits, Each group contains N slits (N is an integer greater than 1), The slits of each group are spaced at a first pitch b1 in a direction perpendicular to the slits, A parallax barrier wherein the groups are spaced in a direction perpendicular to the slit with a second pitch b2 greater than N · b1. The method of claim 1, A parallax barrier with said slits in said each group having substantially the same maximum light transmission. The method of claim 1, A parallax barrier, each of said substantially impermeable regions having a defined width. The method of claim 1, A parallax barrier with said second pitch b2 being substantially equal to 2 · N · b1. The method of claim 1, A parallax barrier wherein N is two. The method of claim 1, A parallax barrier wherein N is three. The method of claim 1, A parallax barrier having an active element having a first and a second mode of operation, said active element providing said group of slits in said first mode of operation and providing an alternative slit arrangement in said second mode. The method of claim 7, wherein A parallax barrier comprising a plurality of parallel slits spaced at substantially uniform pitch in a direction perpendicular to the slits. The method of claim 7, wherein A parallax barrier having a third operating mode in which the active element has an operating region and the active element exhibits a substantially uniform transmission for light throughout the operating region. Including spatial light modulators and parallax barriers, The parallax barrier comprises a plurality of substantially impermeable regions defining a plurality of groups of parallel slits, Each group contains N slits (N is an integer greater than 1), The slits of each group are spaced at a first pitch b1 in a direction perpendicular to the slits, A multi-view display in which groups are spaced apart in a direction perpendicular to the slit at a second pitch b2 greater than Nb1. The method of claim 10, And the modulator comprises a plurality of pixel columns extending parallel to the slots. The method of claim 11, And the columns have a third pitch p in a direction perpendicular to the longitudinal direction of the column, different from the first pitch to provide viewpoint correction. The method of claim 12, The first pitch b1, (Where p is the pitch of the columns and e is the pitch of the main viewing window generated by the display). The method of claim 11, And a third pitch p in a direction perpendicular to the longitudinal direction of the columns, wherein the columns are greater than the first pitch. The method of claim 11, And the columns comprise red, green, and blue columns. The method of claim 15, A multiple view display wherein N is two. The method of claim 16, And the columns arranged as repetitive groups, each group arranged in blue, red, blue, red, green, blue, green, blue, red, green, red and green order. The method of claim 16, And the columns are arranged as repetitive groups, each group arranged in an order of green, green, blue, blue, red and red. The method of claim 15, And wherein N is three. The method of claim 19, Wherein the columns are arranged as eighteen repetitive groups, each group comprising three consecutive pairs of the same triplet, wherein the colors of the triplets of consecutive pairs are cycled by one position relative to each other. The method of claim 19, Wherein the columns are arranged as 36 repetitive groups, each group comprising six consecutive pairs of the same triplet, wherein the triplets of consecutive pairs comprise all permutations of red, green, and blue. The method of claim 10, And a display driver for supplying the modulator with image signals representing a plurality of views as interlaced columns. The method of claim 22, A multi-view display in which the image signals represent two views. The method of claim 22, And a multiple parallax perceptual display in which said image signals represent at least a pair of binocular parallax views.